November 19th, 2019 18:47 -19:30 PDT
Some customers experienced high wait times running tests in our PC, Mac, & Android clouds in our US-West datacenter.
A hypervisor rebooted which caused a load balancer running on it to lose some configuration changes and become overloaded
We restored the configuration values to the proper setting and it recovered.
We have reviewed all systems with high connections loads and verified that their configurations are correct. We have also verified that these changes are in the proper deployment configurations and that they are persistent in all required systems.